Re: [xml] parsing output from xmlNodeDump()

[Daniel Veillard <veillard redhat com>]

On Wed, Sep 08, 2004 at 04:52:18PM -0400, Cory Virok wrote:
> Hello all.
>
> Is there a built in way to parse text that contains entity refs for XML 
> markup without manually search-and-replacing entities? I need to reparse 
> text that has been output by xmlNodeDump().
>
> In my xml document I am storing xml subtrees as encrypted text within 
> XML text nodes. My original XML doc looks like this:
>
> <?xml ...?>
> <root>
>    <secret-stuff>
>       <secret-xml>I'm gonna get encrypted!</secret-xml>
>    </secret-stuff>
> </root>
>
> after encrypting the secret-xml content and making the secret-xml 
> subtree a text node via xmlNodeDump() I get something like this:
>
> <?xml ...?>
> <root>
>    <secret-stuff>
>       &lt;secret-xml&gt;ENCRYPTED_TEXT&lt;/secret-xml&gt;
>    </secret-stuff>
> </root>
>
> Now my question is... is there a built in way to parse the text node so 
> that the entity refs are automagically recognized as XML markup?
> I realize I can just do the search and replace myself... but I'd like to 
> use libXML to do this if I can.

  Seems to me you're mostly making the mistake of dumping <secret-xml>
itself and it subtree instead of dumping its *content* in the first place.
There is now (recently added) xmlParseInNodeContext() which I think
does what you want, but again it's needed only because you xmlNodeDump()
the wrong node !

Daniel

-- 
Daniel Veillard      | Red Hat Desktop team http://redhat.com/
veillard redhat com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/